[PATCH] drm/i915: Do not read non-existent DPLL registers on PCH hardware

[PATCH] drm/i915: Do not read non-existent DPLL registers on PCH hardware

[Chris Wilson]

We only execute intel_decrease_pllclock for pre-PCH hardware, typically
gen4 mobiles. However, in the variable declaration we did read from the
non-PCH DPLL register, quite naughty and detected by SandyBridge.

Reported-by: Andrey Rahmatullin <wrar@wrar.name>
References: https://bugs.freedesktop.org/show_bug.cgi?id=49025
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/intel_display.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index e1716be..10fefdd 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -5486,9 +5486,6 @@ static void intel_decrease_pllclock(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int pipe = intel_crtc->pipe;
-	int dpll_reg = DPLL(pipe);
-	int dpll = I915_READ(dpll_reg);
 
 	if (HAS_PCH_SPLIT(dev))
 		return;
@@ -5501,10 +5498,15 @@ static void intel_decrease_pllclock(struct drm_crtc *crtc)
 	 * the manual case.
 	 */
 	if (!HAS_PIPE_CXSR(dev) && intel_crtc->lowfreq_avail) {
+		int pipe = intel_crtc->pipe;
+		int dpll_reg = DPLL(pipe);
+		u32 dpll;
+
 		DRM_DEBUG_DRIVER("downclocking LVDS\n");
 
 		assert_panel_unlocked(dev_priv, pipe);
 
+		dpll = I915_READ(dpll_reg);
 		dpll |= DISPLAY_RATE_SELECT_FPA1;
 		I915_WRITE(dpll_reg, dpll);
 		intel_wait_for_vblank(dev, pipe);
@@ -5512,7 +5514,6 @@ static void intel_decrease_pllclock(struct drm_crtc *crtc)
 		if (!(dpll & DISPLAY_RATE_SELECT_FPA1))
 			DRM_DEBUG_DRIVER("failed to downclock LVDS!\n");
 	}
-
 }
 
 /**
-- 
1.7.10

Re: [PATCH] drm/i915: Do not read non-existent DPLL registers on PCH hardware

[Daniel Vetter]

On Wed, May 02, 2012 at 12:07:06PM +0100, Chris Wilson wrote:
> We only execute intel_decrease_pllclock for pre-PCH hardware, typically
> gen4 mobiles. However, in the variable declaration we did read from the
> non-PCH DPLL register, quite naughty and detected by SandyBridge.
> 
> Reported-by: Andrey Rahmatullin <wrar@wrar.name>
> References: https://bugs.freedesktop.org/show_bug.cgi?id=49025
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Picked up for -fixes (with Andrey's tested-by added), thanks for the patch.
-Daniel
-- 
Daniel Vetter
Mail: daniel@ffwll.ch
Mobile: +41 (0)79 365 57 48